METEOROLOGY AND DEEP TEMPERATURES 463 



to ascertaining the temperature of the water at the bottom, also 

 noting that at the surface. The temperature at the bottom he 

 measured by means of a slow-action thermometer the construction 

 of which was simple and efficient and the results obtained by its 

 use extremely accurate. For use in lakes he had a thermometer 

 provided with a thick bulb covered with non-conducting material 

 and entirely enclosed in a clear glass bottle full of water. This 

 was lowered to the bottom and left for several hours so that the 

 whole should have acquired the temperature of the lake water. 

 Then it was rapidly raised to the surface and the thermometer 

 read through the sides of the bottle without opening it. Except 

 for the slowness of the process, it is perfect and could not be 

 improved upon. 



De Saussure found that however high the surface temperature 

 might be in summer, the temperature at the bottom of the deep 

 lakes was always within a degree or so of 4 R. (5 C.). The cause 

 of this low temperature puzzled him completely. If he had taken 

 temperature observations at intermediate depths and at different 

 seasons of the year, he would almost certainly have discovered the 

 cause, which is the cooling action at the surface in winter until the 

 temperature there falls to about 4 C., the temperature of maximum 

 density of water. Down to that temperature the surface water 

 sinks, carrying the low temperature to the bottom, but further 

 cooling makes the cold water lighter, and it remains at the surface 

 until it freezes. This normal process is of course subject to 

 variations on account of the mixture of the water by currents due 

 to wind. De Saussure appears to have been ignorant of the fact 

 that water had a maximum density temperature, and the only 

 cause he could assign for the cold deep water was the inflow below 

 the surface of streams fed by melting snow. This explanation 

 he could not accept, but gave it up reluctantly, bringing forward 

 all the arguments against it in great detail, and concluding the 

 whole matter with : ' En attendant, je crois pouvoir affirmer qu'il 

 n'y a aucun principe generalement reconnu qui puisse rendre une 

 raison satisfaisante du froid de nos lacs.' 



Perhaps de Saussure was hindered in his reasoning by includ- 

 ing in the same discussion the phenomena of cold air issuing from 

 the ground in certain localities even in summer. This he accounted 

 for by the cooling by evaporation of water in the passages by which 

 air stored in caverns found its way to the surface. This explana- 

 tion he confirmed by experiment ; but, curiously enough, he 

 imagined that the cooling from this cause could not lower the 

 temperature of air by more than 3 R., one of the few instances of 

 failure to interpret aright the results of his experiments. In this 



